A potential role for interleukin-33 and ?-epithelium sodium channel in the pathogenesis of human malaria associated lung injury

Among clinical parameters, no manifestations of severe falciparum malaria were associated
with a higher incidence of PE, whilst only features which were associated with lower
incidence of PE were anaemia and acidosis. Peripheral parasitaemia on admission did
not differ between PE and non-PE patients, suggesting that multi-organ dysfunction
in this group of patients was not related to greater parasite burden, in agreement
with the study of Gachot et al. 13]. Quantitation of PRBC sequestration in the lung is a better correlate of tissue parasite
burden but again this was not significantly different between cases with or without
PE. CM and metabolic acidosis are the most common manifestations of severe malaria
in all age groups 39]. Severe anaemia, hypoglycaemia and convulsions are more common in semi-immune paediatric
cases (typically in sub-Saharan Africa), whereas renal failure, severe jaundice, ALI,
and ARDS are more frequent in adults 2], 3]. The present data showed a similar distribution of clinical manifestations in both
adults and children (ranging from three to 54 years old), but confirms the clinical
and pathological occurrence of ALI in paediatric cases of severe malaria in Thailand.
However, all of the specimens included in this study were historical samples, all
clinical details were particularly not complete to obtain. Although anti-malarial
treatment was provided in Table 2, there were no available results of blood culture examination and supportive treatments
history, such as pulmonary ventilation. These are the study’s limitation since septicaemia
and tracheal intubation might also be a cause of PE and lowered ?-ENaC 40].

Pathological features of lung injury in the different groups were heterogeneous. Although
PE in severe malaria is characteristically patchy histologically, and we only examined
one to two blocks per case, clinical findings and a threshold pathological score (25 %
of the section showing free fluid in the alveolar space) were used as important criteria
to differentiate PE and non-PE in this study. Some patients dying of malaria show
minimal reaction even to high levels of PRBC and WBC sequestration, through PE to
pneumonitis-like leukocyte infiltration of alveolar capillaries (which may show up
on CXR as patchy infiltrates and be misdiagnosed as PE) to full blown ARDS. Ten patients
in this study showed evidence of hyaline membrane formation evidence of early diffuse
alveolar damage (DAD), which is correlated with the clinical development of ARDS.
PE is likely to be an early event and could be influenced by direct adhesive signalling
events resulting from sequestration of PRBC and WBC, and immune-inflammatory network
activation in reaction to that in alveolar capillaries. However other factors such
as disturbance of epithelial function may also be important, hence examining IL-33,
sodium channel and AQPs.

Bronchial expression of IL-33 was significantly increased in patients with PE compared
to those without PE and healthy subjects. Furthermore, a positive correlation was
found between bronchial IL-33 expression and the number of CD68 + monocytes and neutrophils,
septal congestion score, WBC score, and hyaline membrane formation score, suggesting
IL-33 may play-role in lung destruction in severe malaria. Some studies showed that
IL-33 may be elaborated T cells tracking to an inflammatory site, served as chemo-attractant
for human T helper (Th)2 cells, a crucial cytokine for Th2-mediated host response,
and a central role in epithelial cells controlling immune responses 41], 42]. It is able to activate cells of both the innate and adaptive immune system, and
depending on the disease can either promote the resolution of inflammation or drive
disease pathology 35]. According to previous studies, falciparum malaria infection is characterized by
elevation of Th2 cytokines expression 43], 44]. However the mechanism by which IL-33 is involved in lung injury in falciparum malaria
remains unclear, thorough analysis of either Th1 and Th2 cytokines or Th cell sub-sets
with their relevant transcription factors contributing to the level of bronchial epithelial
IL-33 still needs to be further studied.

In the lung, IL-33 is mainly expressed in bronchial epithelial cells 45]. Thus, airway epithelial cells are also active players in the pathogenesis of lung
malaria through epithelial cytokines, including IL-33 produced in response to various
exogenous stimuli or to cellular damage 46]. IL-33 is elevated in patients with pulmonary diseases 35], 47]. A recent report on IL-33 transgenic mice showed pulmonary inflammation with a significant
accumulation of lymphocytes, monocytes and neutrophils but mild accumulation of eosinophils
in the lung 47], 48]. This observation may correspond to a result of the significant positive correlation
between bronchial IL-33 and CD68, neutrophil and WBCs score. Although the results
showed a negative correlation between bronchial IL-33 and CD8 T cells, in P. falciparum infection, CD8 + cytotoxic T lymphocytes were significantly reduced indicating that
immune suppression is more pronounced 44].

Patients with severe malaria due to P. falciparum, and more rarely due to P. vivax and Plasmodium knowlesi may develop ALI and ARDS, often several days after anti-malarial drug treatment 2], 3]. The result demonstrated a significant positive correlation between bronchial IL-33
and hyaline membrane formation, WBC and neutrophils together with a significantly
higher incidence of hyaline membrane formation in PE than non-PE. These findings are
in agreement with the notion that ARDS pathophysiology centres on inflammatory-mediated
increased capillary permeability or endothelial damage, leading to DAD that can continue
after parasite clearance 2].

Although this study demonstrated a similar range of leukocyte sub-sets (CD3, CD8,
CD68, and neutrophil) among PE, non-PE and healthy subjects, the total score of WBC
accumulation was significantly higher in severe malaria patients compared to normal
cases. This supports the concept of excessive accumulation of WBCs in the lung as
a feature contributing to severe malaria ALI. Interestingly, the correlation between
bronchial IL-33 and WBCs was found as previously described. These also appeared in
a previous study on paediatric CM 49] which indicated that the number of pulmonary T-cells (as measured by CD3, CD4 or
CD8) and pulmonary macrophages (CD68) were similar across CM, non-CM, and non-malarial
diagnosed cases.

The results demonstrate that the VSMC expression of ?-ENaC was significantly decreased
in patients with PE compared with non-PE and healthy subjects. Moreover, bronchial
expression of ?-ENaC was significantly decreased in patients with and without PE when
compared to healthy subjects. Due to the age of some of the cases, and prolonged storage,
a constitutive antigen was examined to compare the potential loss of antigenicity
in sections. However AE1/AE3 is a board spectrum pancytokeratin stain and quantitatively
showed no differences in lung of patients with or without PE, and healthy subjects.
Similarly, the expression of AQP-1 and -5 showed no significant differences amongst
malarial patients with or without PE, or between malaria cases and controls.

There was a significant negative correlation between VSMC and bronchial ?-ENaC expression
and the percentage of PRBC sequestration, alveolar thickness, alveolar expansion score,
septal congestion score, and malarial pigment score, suggesting that ?-ENaC may play
an important role in mediating clearance of oedema fluid from the lung in severe malaria.

The importance of ?-ENaC protein in VSMC-mediated, pressure-induced constriction in
arteries has been demonstrated 50], 51]. The myogenic response is an important regulatory mechanism for blood flow autoregulation
to the circulation 52]. The reduction of VSMC ?-ENaC leads to decreased myogenic constriction and consequently
to increased pressure transmission to the delicate microvasculature and increased
susceptibility to end organ injury 53]. In the current study, VSMC ?-ENaC was significantly decreased in patients with PE.
Lung injury in PE may result from an increased lung microvascular pressure due to
downregulation of VSMC ?-ENaC, leading to an increase permeability of the gas-blood
barrier, which is likely to contribute to the formation or maintenance of either interstitial
or alveolar PE.

The bronchial circulation also plays a significant role in the formation and re-absorption
of both hydrostatic and permeability oedema 54] mediated by various ion pumps and channels such as AQPs, cystic fibrosis transmembrane
conductance regulator (CFTR), ENaC, and cyclic nucleotide-gated (CNG) 55]. Severe malaria patients with PE had a reduced bronchial ?-ENaC expression. In addition,
the low expression level of both VSMC and bronchial ?-ENaC was correlated to high
percentage of alveolar vascular sequestration of PRBC and high malarial pigment scores,
implying a high tissue parasite burden in these patients. The mechanisms by which
pathogens affect ENaC expression are still being elucidated; C57BL/6 mice infected
with Plasmodium berghei K173 (PbK) demonstrated a reduction of activity and expression of ENaC suggesting
that the infectious agents can have direct effects on ENaC in the lung epithelium
19]. These results imply that falciparum malaria causes a down regulation of ?-ENaC contributing
to PE.

Hypoxia has been shown to reduce ENaC activity 56]. The results demonstrate that severe malaria patients who presented anaemia and pneumonia
had reduced expression of bronchial ?-ENaC, and these clinical complications can lead
to hypoxemic condition and alveolar hypoxia, respectively.

In severe malaria patients, no difference was seen in the expression of AQP-1 and
-5 between controls and malaria cases either with or without PE. This is in agreement
with many reports showing that AQP-1 and -5 does not appear to play a role in the
physiological clearance of oedema in the lung or in the accumulation of extravascular
water in the injured lung 27]–29], 57]. These studies suggest that AQP-1 and -5 only provide a pathway for osmotically induced
water transport across the pleural barrier and do not affect pleural fluid dynamics
in the endothelial cell injury.

These results suggest that induced bronchial IL-33 expression may be involved in lung
injury and the genesis of PE in severe malaria ALI. Moreover, the reduction both VSMC
and bronchial ?-ENaC may retard fluid clearance from the lung and lead to PE in severe
malaria. Alveolar AQP-1 and -5 did not show changes in expression, so do not appear
to play a major role in deregulated water movement in malaria-induced lung injury.
Results from this study identify a potential casual relationship between ?-ENaC and
lung oedema pathogenesis in severe malaria. However other aspects require further
study, for example the mechanism of channel gating by either mechanical stimuli or
extrinsic factors. Not all patients with severe malaria develop lung injury and in
this study, histopathological features were broadly similar between groups exhibiting
PE or not, hence more detailed research is required to determine the factors which
lead to PE or the development of more severe ARDS in individual patients.